Tuberculous, brucellar and pyogenic spondylitis: comparison of magnetic resonance imaging findings and assessment of its value.

To the Editor: Infectious spondylitis is an infection by a specific organism of one or more components of the spine, namely the vertebrae, intervertebral disks, paraspinal abnormal soft tissues, and epidural space. Differential diagnosis enables proper treatment of the different types and may reduce the rate of disability and functional impairment. Differentiating the types of spondylitis clinically and radiographically is challenging. (1) We aimed to assess the value of magnetic resonance (MR) imaging in the differentiation of tuberculous (TB), brucellar and pyogenic spondylitis.

The MR images of 28 patients with TB spondylitis (17 male/11 female), 21 patients with brucellar spondylitis (13 male/8 female), and 26 (16 male/10 female) patients with pyogenic spondylitis were analyzed without knowing clinical features or pathology results. The diagnoses of TB and pyogenic spondylitis were confirmed by biopsy (histologic and microbiologic studies). The diagnoses of brucellar spondylitis were established by a positive standard tube agglutination test ([greater than or equal to]1/160) and/or a positive blood culture. The mean interval from presentation to MRI was 38 weeks in TB, 30 weeks in brucellar spondylitis and 15 weeks in pyogenic spondylitis. MR imaging with gadopentetate dimeglumine was obtained in all patients.

The reviewers evaluated the presence and margins of paraspinal abnormal signal, paraspinal and psoas abscess, the extent of subligamentous spread, the number of involved vertebral bodies, the involvement site of the spine (cervical-dorsal-lumbar-sacral), the affected area of the body (anterior-posterior, corpus-endplate), diffuse, focal, or patchy body involvement, the signal intensity of the involved vertebral bodies (iso, hypo, or hyperintense) and disks, the amount of paraspinal and epidural abnormal soft tissue, the presence of spinal deformity and intraosseous abscess, loss of vertebral contour, as well as the presence of multicompartmental and IM abscess.

The patients were divided into four groups according to the amount of paraspinal and epidural abnormal soft tissue (category 0: no soft tissue, category 1: less than 5 mm, category 2: between 5 and 10 mm, category 3: more than 10 mm) and compared.

In TB spondylitis, thoracic involvement was observed in 52/101 (51%) of total vertebrae involved and lumbar involvement in 24/101 (24%). These figures were 15/54 (28%) and 24/54 (44%) in brucellosis; 9/62 (15%) and 47/62 (76%) in pyogenic spondylitis, respectively. Well-defined paraspinal signals were detected in most of the TB (25/27; 93%) and brucellosis (17/20; 85%) cases, whereas ill-defined signals were seen in pyogenic spondylitis (14/26; 54%), suggesting that this finding may be used to differentiate pyogenic spondylitis from the others.

Both anterior and posterior vertebrae were involved in 16/28 (57%) of TB and in 20/26 (77%) of pyogenic patients, but in only 19% of brucellosis cases. The anterior vertebra alone was involved in 17/21 (81%) of brucellosis, as opposed to 43% of TB and 23% of pyogenic patients. Diffuse vertebral involvement was observed in most of the cases in all groups. Subligamentous spread was detected in 23/28 (82%) and spinal deformity was detected in 10/28 (36%) of TB, which is consistent with previous reports. (2-3) Spinal deformity is closely related with the destructive nature of caseating granuloma and is an important diagnostic clue when present. (4) Paraspinal and intraosseous abscess, subligamentous extension and spinal deformity were detected more frequently in TB than in others.

There were no significant differences in the number of vertebrae involved, sacral involvement, presence of IM abscess, loss of vertebral contour, the pattern of vertebral involvement, involvement of intervertebral disk and disk space narrowing.

Generally, the size of the paraspinal abnormal soft tissue is larger in TB than that of the other infections. (5) In our study, larger amounts of paraspinal abnormal soft tissue were mainly found in category 3 in TB (20/28; 71%); category 2 in brucellosis (13/21; 62%); and in both category 2 (10/26; 38%) and category 3 (12/26; 46%) in pyogenic cases. We found significant differences between the groups in regards to the amount of paraspinal abnormal soft tissue but not in the epidural.

Our study revealed that distinctive MR features of TB spondylitis include extensive paraspinal abnormal soft tissue (>1 cm), predilection for thoracic involvement, well-defined paraspinal abnormal signal, subligamentous spread and severe spinal deformities. Those of brucellar spondylitis include predilection for lumbar spine, moderate paraspinal abnormal soft tissue, well-defined paraspinal abnormal signal, diffuse, but solely anterior part involvement, and no spinal deformity. Distinctive MR features of pyogenic spondylitis include moderate paraspinal abnormal soft tissue, ill-defined paraspinal signal, and anterior-posterior vertebral involvement without gibbus deformity. We conclude that these MRI findings are distinct enough to differentiate TB, brucellar and pyogenic spondylitis.

Mehmet Halit Yilmaz, MD

Department of Radiology

Bilgul Mete, MD

Department of Infectious Diseases and Clinical Microbiology

Fatih Kantarci, MD

Department of Radiology

Resat Ozaras, MD

Department of Infectious Diseases and Clinical Microbiology

Harun Ozer, MD

Department of Radiology

Ali Mert, MD

Department of Infectious Diseases and Clinical Microbiology

Ismail Mihmanli, MD

Department of Radiology

Recep Ozturk, MD

Department of Infectious Diseases and Clinical Microbiology

Kaya Kanberoglu, MD

Department of Radiology

Cerrahpasa Medical Faculty, Istanbul University

Istanbul, Turkey

References

1. Arizono T, Oga M, Shiota E, et al. Differentiation of vertebral osteomyelitis and tuberculous spondylitis by magnetic resonance imaging. Int Orthop 1995;19:319-322.

2. Colmenero JD, Jimenez-Mejias ME, Sanchez-Lora FJ, et al. Pyogenic, tuberculous, and brucellar vertebral osteomyelitis: a descriptive and comparative study of 219 cases. Ann Rheum Dis 1997;56:709-715.

3. Jung NY, Jee WH, Ha KY, et al. Discrimination of tuberculous spondylitis from pyogenic spondylitis on MRI. AJR Am J Roentgenol 2004;182:1405-1410.

4. Buchelt M, Lack W, Kutschera HP, et al. Comparison of tuberculous and pyogenic spondylitis. An analysis of 122 cases. Clin Orthop Relat Res 1993;296:192-199.

5. Sharif HS, Aideyan OA, Clark DC, et al. Brucellar and tuberculous spondylitis: comparative imaging features. Radiology 1989;171:419-425.